If there’s a version of the ‘tastes great, less filling’ argument in the Hadoop marketplace, it’s probably about whether Hadoop’s value is in being cost-effective vs. delivering ‘strategic’ ROI. The financial services industry is a particularly interesting domain in which to work out the balance of assets and liabilities.
One especially useful answer to this question (like much great beer) comes from Canada – the Royal Bank of Canada (RBC) to be specific, in the person of Reid Levesque, who spoke on ‘Beyond TCO: Architecting Hadoop for Adoption and Data Applications‘ at the most recent Hadoop Summit in San Jose.
What’s interesting is the approach that Mr. Levesque took to TCO: recognizing Hadoop’s merits, his team worked through a series of use cases. Consider the following multiple choice list:
- [a] File Archive … + Hadoop
- [b] Data Intensive Grid Compute Analytics … + Hadoop
- [c] Database Replacement … + Hadoop
- [d] ETL Offload … + Hadoop
- [e] None of the above
- [f] All of the above
Don’t turn your browser upside down to look for the answers below. They’re in the video. But here’s the one line summary:
“We arrived at a use case that was only possible with Hadoop, but would appeal to a broad audience of users who were not principally data scientists nor software developers.”
After watching the video, you may want to read up on the value of self-service big data.